1 Nez Perce Tribe, Department of Fisheries Resources Management, 14054 Burr Dr., McCall, Idaho, 83638, USA

✉ Correspondence: Jay Hesse <>, Ryan N. Kinzer <>

DRAFT

1 Quasi-Extinction Summary

Updates for the 2024 analysis:

  • Removed Little Salmon River due to limited data from only a small portion of the population.
  • Included multiple time-series of observations for each population if available (i.e, spawning ground survey data, weir observations, PIT-tag based estimates)
  • Performed model selection on multiple candidate models with varying numbers of state processes, process error, and observation error parameters

1.1 Spring/summer Chinook Salmon

Highlights:

  • Seven state processes and trends, aligned with major population groups, represent spring/summer Chinook Salmon populations
  • Spring/summer Chinook Salmon populations have declined by an average of 6% annually for the last 10-years
  • All populations below critical abundance thresholds
    • 4 populations currently below the QET50
    • 8 populations below 50 in 2024
    • 2 populations predicted to fall below 50 by 2029
    • 20 populations remain above 50 in 2029
Current status of natural-origin Snake River spring/summer Chinook Salmon relative to the quasi-extinction threshold (QET) and Columbia Basin Partnership goals.

Figure 1.1: Current status of natural-origin Snake River spring/summer Chinook Salmon relative to the quasi-extinction threshold (QET) and Columbia Basin Partnership goals.

Estimated slope parameters for natural-origin Snake River spring/summer Chinook Salmon abundance trends indicate an average annual decline of 6% for the last 10-years.

Figure 1.2: Estimated slope parameters for natural-origin Snake River spring/summer Chinook Salmon abundance trends indicate an average annual decline of 6% for the last 10-years.

1.2 Summer Steelhead

Highlights:

  • A single state process and trend represents all summer steelhead populations
  • Summer steelhead have declined by 11% annually for the last 10-years
  • All populations below critical abundance thresholds
    • 3 populations currently below the QET50
    • 3 populations predicted to fall below 50 by 2029
    • 16 populations remain above 50 in 2029
Current status of natural-origin Snake River summer steelhead relative to the quasi-extinction threshold (QET) and Columbia Basin Partnership goals.

Figure 1.3: Current status of natural-origin Snake River summer steelhead relative to the quasi-extinction threshold (QET) and Columbia Basin Partnership goals.

Modeled abundance trends of natural-origin Snake River summer steelhead indicate an annual 11% decline for the last 10-years.

Figure 1.4: Modeled abundance trends of natural-origin Snake River summer steelhead indicate an annual 11% decline for the last 10-years.

2 Methods

2.1 Datasets

  1. NOSAij: derived by co-managers and downloaded from Coordinated Assessments on 3/05/2025
    • Generally estimated from expanded redd counts or mark/recapture weir estimates
  2. Escapement: derived from PIT-tag observations and a branch occupancy model similar to Waterhouse et al. 2020
    • includes a time-varying component to better describe run-timings of different populations passing Lower Granite Dam

2.2 Data Analysis

2.2.1 Multivariate Auto-Regressive State Space Model (Holmes et al. 2012)

\[\begin{aligned} \textbf{State process:} \quad & \mathbf{x}_{t} = \mathbf{x}_{t-1} + \mathbf{u} + \mathbf{w}_t, \quad \mathbf{w}_t \sim \mathcal{N}(0, \mathbf{Q}) \\ \textbf{Observation process:} \quad & \mathbf{y}_{t} = \mathbf{Z} \mathbf{x}_{t} + \mathbf{a} + \mathbf{v}_t, \quad \mathbf{v}_t \sim \mathcal{N}(0, \mathbf{R}) \end{aligned}\]
  • \(\mathbf{x}_t\): vector of hidden state values at time \(t\)
  • \(\mathbf{u}\): state intercept (drift) vector
  • \(\mathbf{y}_t\): vector of observed values at time \(t\)
  • \(\mathbf{Z}\): observation matrix mapping states to observations
  • \(\mathbf{a}\): observation bias vector

2.2.2 Fitted Models

  • State process ( \(\mathbf{u}\) ):
    • A single basin-wide process (1)
    • A process for each major population group (7 - sp/sm Chinook Salmon and 5 - summer steelhead)
    • A process for each population (34 - sp/sm Chinook Salmon and 22 summer steelhead)
  • Process Error ( \(\mathbf{Q}\) )
    • Equal Variance-Covariance (2)
    • Diagaonl and Equal (1)
    • Diagonal and Unequal (equals number of state processes)
  • Observation Error ( \(\mathbf{R}\) )
    • Diagonal and Equal (1)
    • Diagonal and Unequal (equals number of observation time-series)

3 Results

3.1 Spring/summer Chinook Salmon

3.1.1 Observations

Table 3.1: The method of time-series observations used for Snake River spring/summer Chinook Salmon abundance modeling differed across populations with the first observation beginning in 1980 and last observation in 2024.
MPG Population Method First Last n min max average sd
Dry Clearwater Upper South Fork Clearwater PIT-tag Observations 2012 2024 12 75 1120 408 327
Grande Ronde / Imnaha Big Sheep Creek PIT-tag Observations 2011 2024 13 9 340 77 86
Catherine Creek SGS and Weir Expansions 1980 2024 45 0 1034 210 229
PIT-tag Observations 2015 2024 9 79 430 183 123
Grande Ronde River Upper Mainstem SGS and Weir Expansions 1980 2024 45 0 545 89 101
PIT-tag Observations 2018 2024 6 13 93 40 29
Imnaha River Mainstem SGS and Weir Expansions 1980 2024 45 139 1465 515 357
PIT-tag Observations 2011 2024 13 176 1731 605 437
Lookingglass Creek PIT-tag Observations 2010 2024 14 33 288 117 84
Minam River SGS and Weir Expansions 1980 2024 45 30 1209 396 252
Wallowa/Lostine Rivers SGS and Weir Expansions 1980 2023 44 37 1283 409 342
Wenaha River SGS and Weir Expansions 1980 2024 45 30 767 333 217
PIT-tag Observations 2019 2024 5 118 577 218 201
Lower Snake Asotin Creek SGS and Weir Expansions 1984 2016 33 0 35 7 9
PIT-tag Observations 2010 2024 14 0 312 53 85
Tucannon River SGS and Weir Expansions 1980 2024 45 4 852 270 232
Middle Fork Salmon River Bear Valley Creek SGS and Weir Expansions 1980 2024 45 17 1315 382 334
PIT-tag Observations 2015 2024 9 6 1422 327 443
Big Creek SGS and Weir Expansions 1980 2024 45 4 682 181 158
PIT-tag Observations 2012 2024 12 140 1147 619 338
Camas Creek SGS and Weir Expansions 1980 2024 44 0 294 64 70
Chamberlain Creek SGS and Weir Expansions 1985 2024 36 0 711 223 179
Loon Creek SGS and Weir Expansions 1980 2024 44 0 654 85 135
Marsh Creek SGS and Weir Expansions 1980 2024 45 0 872 251 222
Middle Fork Salmon River Lower Mainstem SGS and Weir Expansions 1987 2024 37 0 63 7 14
Middle Fork Salmon River Upper Mainstem SGS and Weir Expansions 1995 2024 30 1 322 84 78
Sulphur Creek SGS and Weir Expansions 1980 2024 45 0 303 68 71
South Fork Salmon River East Fork South Fork Salmon River SGS and Weir Expansions 1987 2024 38 5 844 251 199
PIT-tag Observations 2010 2024 14 191 1223 640 323
Secesh River SGS and Weir Expansions 1996 2024 29 139 1460 555 370
PIT-tag Observations 2010 2024 14 213 1538 735 395
South Fork Salmon River SGS and Weir Expansions 1980 2024 45 64 1797 604 404
PIT-tag Observations 2010 2024 14 157 3583 1186 1034
Upper Salmon River East Fork Salmon River SGS and Weir Expansions 1980 2024 45 7 866 278 242
Lemhi River SGS and Weir Expansions 1980 2024 45 10 708 194 163
PIT-tag Observations 2010 2024 14 87 745 324 216
North Fork Salmon River SGS and Weir Expansions 1991 2024 34 2 324 73 73
PIT-tag Observations 2016 2024 7 23 358 106 115
Pahsimeroi River SGS and Weir Expansions 1980 2024 37 0 559 168 142
Panther Creek PIT-tag Observations 2018 2024 6 82 371 187 107
Salmon River Lower Mainstem SGS and Weir Expansions 1980 2024 45 11 449 120 99
Salmon River Upper Mainstem SGS and Weir Expansions 1980 2024 45 23 991 365 242
Valley Creek SGS and Weir Expansions 1980 2024 45 0 325 102 88
PIT-tag Observations 2010 2024 14 79 675 275 185
Yankee Fork SGS and Weir Expansions 1980 2024 44 0 358 56 85
PIT-tag Observations 2012 2024 12 28 296 110 99
Wet Clearwater Lochsa River PIT-tag Observations 2017 2024 7 135 573 292 149
Lolo Creek PIT-tag Observations 2012 2024 12 34 312 125 98

3.1.2 Model Fits

Table 3.2: Candidate models and number of parameters fit to spring/summer Chinook Salmon time-series observations to explain Snake River abundance trends.
Model Id Total Parameters U Q R logLik AICc \(\Delta\)AIC
4 106 7 2 48 -1409.39 3048.45 0.00
2 99 1 1 48 -1466.43 3146.20 97.75
10 133 34 2 48 -1434.48 3163.32 114.87
8 111 7 7 48 -1483.20 3207.85 159.39
6 105 7 1 48 -1497.10 3221.52 173.07
3 59 7 2 1 -1592.07 3307.45 259.00
1 52 1 1 1 -1644.39 3396.90 348.45
9 86 34 2 1 -1619.06 3421.60 373.15
7 64 7 7 1 -1658.74 3451.76 403.30
5 58 7 1 1 -1675.38 3471.89 423.44

3.1.3 State Process

Estimated abundance (natural-log) trends for the seven state processes (xtT) estimated from the best fitting Snake River spring/summer Chinook Salmon model (grey shading represents 95% CI's).

Figure 3.1: Estimated abundance (natural-log) trends for the seven state processes (xtT) estimated from the best fitting Snake River spring/summer Chinook Salmon model (grey shading represents 95% CI’s).

3.1.4 Modeled Observations

Empirical natural-origin abundance estimates for Snake River spring/summer Chinook Salmon (points) and estimated population trends (ytT) from the best fitting model (line).

Figure 3.2: Empirical natural-origin abundance estimates for Snake River spring/summer Chinook Salmon (points) and estimated population trends (ytT) from the best fitting model (line).

3.1.5 5-Year Predictions

Estimated and predicted population abundance trends for natural-origin Snake River Chinook Salmon. Red points and triangles indicated an estimated annual return of 50 or fewer spawners.

Figure 3.3: Estimated and predicted population abundance trends for natural-origin Snake River Chinook Salmon. Red points and triangles indicated an estimated annual return of 50 or fewer spawners.

3.2 Summer Steelhead

3.2.1 Observations

Table 3.3: The method of time-series observations used for Snake River summer steelhead abundance modeling were primarily based on PIT-tag observations with the first beginning in 2010 and the last observation in 2024. Three populations included a secondary time-series based on spawning ground survey expansions.
MPG Population Method First Last n min max average sd
Clearwater River Clearwater River Lower Mainstem PIT-tag Observations 2010 2024 15 34 995 404 299
Lochsa River PIT-tag Observations 2017 2024 8 154 1263 562 350
Lolo Creek PIT-tag Observations 2012 2024 13 65 657 274 194
Selway River PIT-tag Observations 2017 2024 8 158 907 475 267
South Fork Clearwater River PIT-tag Observations 2012 2024 13 113 1232 500 376
Grande Ronde River Grande Ronde River Lower Mainstem PIT-tag Observations 2019 2023 5 247 493 383 95
Grande Ronde River Upper Mainstem PIT-tag Observations 2013 2024 12 223 2295 797 584
SGS and Weir Expansions 1980 2018 39 341 6461 1884 1270
Joseph Creek PIT-tag Observations 2011 2024 14 281 3202 1197 890
SGS and Weir Expansions 1980 2017 38 574 6476 2489 1658
Wallowa River PIT-tag Observations 2014 2024 11 287 1020 601 256
Imnaha River Imnaha River PIT-tag Observations 2011 2024 14 602 3408 1607 960
Lower Snake Asotin Creek PIT-tag Observations 2010 2024 15 166 1945 672 548
Tucannon River PIT-tag Observations 2010 2024 15 194 981 485 240
SGS and Weir Expansions 2010 2023 14 316 1362 714 360
Salmon River East Fork Salmon River PIT-tag Observations 2011 2019 8 0 59 18 21
Lemhi River PIT-tag Observations 2010 2024 15 38 563 239 176
Middle Fork Salmon River Lower Mainstem PIT-tag Observations 2011 2024 14 68 860 295 233
Middle Fork Salmon River Upper Mainstem PIT-tag Observations 2020 2024 5 14 63 34 18
North Fork Salmon River PIT-tag Observations 2017 2024 7 0 812 159 290
Pahsimeroi River PIT-tag Observations 2011 2024 14 0 157 58 55
Panther Creek PIT-tag Observations 2018 2024 7 82 198 144 42
Salmon River Upper Mainstem PIT-tag Observations 2010 2024 15 28 440 150 121
Secesh River PIT-tag Observations 2010 2024 15 14 359 115 105
South Fork Salmon River PIT-tag Observations 2010 2024 15 75 2050 669 594

3.2.2 Model Fits

Table 3.4: Candidate models and number of parameters fit to summer steelhead time-series observations to explain Snake River abundance trends.
Model Id Total Parameters u Q R logLik AICc \(\Delta\)AIC
2 53 1 1 25 -173.6063 477.5700 0.00000
4 58 5 2 25 -172.4959 490.7482 13.17829
10 75 22 2 25 -152.4930 508.5072 30.93727
6 57 5 1 25 -197.1921 537.0077 59.43771
8 61 5 5 25 -195.7174 546.7564 69.18648
1 29 1 1 1 -290.3841 645.4864 167.91640
12 74 22 1 25 -227.4331 654.7353 177.16530
3 34 5 2 1 -290.2449 657.8599 180.28994
9 51 22 2 1 -279.5091 683.3980 205.82803
7 37 5 5 1 -304.4895 694.1822 216.61226

3.2.3 State Process

Estimated abundance (natural-log) trend for a single state process (xtT) as estimated from the best fitting Snake River summer steelhed model (grey shading represents 95% CI's).

Figure 3.4: Estimated abundance (natural-log) trend for a single state process (xtT) as estimated from the best fitting Snake River summer steelhed model (grey shading represents 95% CI’s).

3.2.4 Modeled Observations

Empirical natural-origin abundance estimates for Snake River summer steelhead (points) and estimated population trends (ytT) from the best fitting model (line).

Figure 3.5: Empirical natural-origin abundance estimates for Snake River summer steelhead (points) and estimated population trends (ytT) from the best fitting model (line).

3.2.5 5-Year Predictions

Estimated and predicted population abundance trends for natural-origin Snake River summer steelhead. Red points and triangles indicated an estimated annual return of 50 or fewer spawners.

Figure 3.6: Estimated and predicted population abundance trends for natural-origin Snake River summer steelhead. Red points and triangles indicated an estimated annual return of 50 or fewer spawners.